CPU-Only & High-Memory Workloads

Not all scientific computations require GPUs. For high-memory genome assembly, chemistry simulations, or heavy data wrangling, Berkelium features scale-up CPU nodes powered by AMD EPYC 9754 (128 cores / 256 threads) with 1.5 TB DDR5 ECC RAM.

Example: High-Memory Simulation Manifest

apiVersion: batch/v1
kind: Job
metadata:
  name: highmem-genome-assembly
  namespace: sci-myproject
spec:
  template:
    spec:
      restartPolicy: OnFailure
      containers:
        - name: assembler
          image: biocontainers/spades:v3.15.5_cv1
          command: ["spades.py", "-1", "/data/reads_R1.fq", "-2", "/data/reads_R2.fq", "-o", "/data/assembly", "-t", "64", "-m", "500"]
          volumeMounts:
            - name: dataset
              mountPath: /data
          resources:
            requests:
              cpu: "32"
              memory: "256Gi"
            limits:
              cpu: "64"
              memory: "512Gi"
      volumes:
        - name: dataset
          persistentVolumeClaim:
            claimName: lab-shared-data-pvc

---

OpenMP & Thread Pinning

When running multi-threaded scientific binaries, ensure OpenMP threads match your requested CPU limits:

export OMP_NUM_THREADS=${OMP_NUM_THREADS:-$(nproc)}
export MKL_NUM_THREADS=$OMP_NUM_THREADS